{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.977220956719817,
  "eval_steps": 500,
  "global_step": 2190,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004555808656036446,
      "grad_norm": 362.0,
      "learning_rate": 9.132420091324201e-07,
      "loss": 46.6393,
      "step": 1
    },
    {
      "epoch": 0.022779043280182234,
      "grad_norm": 372.0,
      "learning_rate": 4.566210045662101e-06,
      "loss": 46.947,
      "step": 5
    },
    {
      "epoch": 0.04555808656036447,
      "grad_norm": 190.0,
      "learning_rate": 9.132420091324201e-06,
      "loss": 43.7966,
      "step": 10
    },
    {
      "epoch": 0.0683371298405467,
      "grad_norm": 128.0,
      "learning_rate": 1.3698630136986302e-05,
      "loss": 35.4793,
      "step": 15
    },
    {
      "epoch": 0.09111617312072894,
      "grad_norm": 40.75,
      "learning_rate": 1.8264840182648402e-05,
      "loss": 28.4066,
      "step": 20
    },
    {
      "epoch": 0.11389521640091116,
      "grad_norm": 19.5,
      "learning_rate": 2.2831050228310503e-05,
      "loss": 24.6114,
      "step": 25
    },
    {
      "epoch": 0.1366742596810934,
      "grad_norm": 15.8125,
      "learning_rate": 2.7397260273972603e-05,
      "loss": 22.96,
      "step": 30
    },
    {
      "epoch": 0.15945330296127563,
      "grad_norm": 9.375,
      "learning_rate": 3.1963470319634704e-05,
      "loss": 21.0704,
      "step": 35
    },
    {
      "epoch": 0.18223234624145787,
      "grad_norm": 4.75,
      "learning_rate": 3.6529680365296805e-05,
      "loss": 19.7729,
      "step": 40
    },
    {
      "epoch": 0.20501138952164008,
      "grad_norm": 4.59375,
      "learning_rate": 4.1095890410958905e-05,
      "loss": 19.4123,
      "step": 45
    },
    {
      "epoch": 0.22779043280182232,
      "grad_norm": 5.125,
      "learning_rate": 4.5662100456621006e-05,
      "loss": 18.5253,
      "step": 50
    },
    {
      "epoch": 0.2505694760820046,
      "grad_norm": 7.25,
      "learning_rate": 5.0228310502283106e-05,
      "loss": 17.9153,
      "step": 55
    },
    {
      "epoch": 0.2733485193621868,
      "grad_norm": 10.9375,
      "learning_rate": 5.479452054794521e-05,
      "loss": 16.7799,
      "step": 60
    },
    {
      "epoch": 0.296127562642369,
      "grad_norm": 18.125,
      "learning_rate": 5.936073059360731e-05,
      "loss": 15.261,
      "step": 65
    },
    {
      "epoch": 0.31890660592255127,
      "grad_norm": 28.875,
      "learning_rate": 6.392694063926941e-05,
      "loss": 11.9226,
      "step": 70
    },
    {
      "epoch": 0.3416856492027335,
      "grad_norm": 27.25,
      "learning_rate": 6.84931506849315e-05,
      "loss": 7.2453,
      "step": 75
    },
    {
      "epoch": 0.36446469248291574,
      "grad_norm": 6.375,
      "learning_rate": 7.305936073059361e-05,
      "loss": 3.4058,
      "step": 80
    },
    {
      "epoch": 0.38724373576309795,
      "grad_norm": 5.84375,
      "learning_rate": 7.76255707762557e-05,
      "loss": 2.3455,
      "step": 85
    },
    {
      "epoch": 0.41002277904328016,
      "grad_norm": 2.28125,
      "learning_rate": 8.219178082191781e-05,
      "loss": 2.0168,
      "step": 90
    },
    {
      "epoch": 0.4328018223234624,
      "grad_norm": 4.5,
      "learning_rate": 8.67579908675799e-05,
      "loss": 1.7945,
      "step": 95
    },
    {
      "epoch": 0.45558086560364464,
      "grad_norm": 0.79296875,
      "learning_rate": 9.132420091324201e-05,
      "loss": 1.6782,
      "step": 100
    },
    {
      "epoch": 0.4783599088838269,
      "grad_norm": 1.3046875,
      "learning_rate": 9.58904109589041e-05,
      "loss": 1.575,
      "step": 105
    },
    {
      "epoch": 0.5011389521640092,
      "grad_norm": 2.015625,
      "learning_rate": 0.00010045662100456621,
      "loss": 1.506,
      "step": 110
    },
    {
      "epoch": 0.5239179954441914,
      "grad_norm": 1.8125,
      "learning_rate": 0.00010502283105022832,
      "loss": 1.47,
      "step": 115
    },
    {
      "epoch": 0.5466970387243736,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00010958904109589041,
      "loss": 1.433,
      "step": 120
    },
    {
      "epoch": 0.5694760820045558,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00011415525114155252,
      "loss": 1.392,
      "step": 125
    },
    {
      "epoch": 0.592255125284738,
      "grad_norm": 0.828125,
      "learning_rate": 0.00011872146118721462,
      "loss": 1.354,
      "step": 130
    },
    {
      "epoch": 0.6150341685649203,
      "grad_norm": 1.921875,
      "learning_rate": 0.0001232876712328767,
      "loss": 1.336,
      "step": 135
    },
    {
      "epoch": 0.6378132118451025,
      "grad_norm": 0.58203125,
      "learning_rate": 0.00012785388127853882,
      "loss": 1.3193,
      "step": 140
    },
    {
      "epoch": 0.6605922551252847,
      "grad_norm": 2.125,
      "learning_rate": 0.00013242009132420092,
      "loss": 1.3012,
      "step": 145
    },
    {
      "epoch": 0.683371298405467,
      "grad_norm": 2.015625,
      "learning_rate": 0.000136986301369863,
      "loss": 1.2792,
      "step": 150
    },
    {
      "epoch": 0.7061503416856492,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001415525114155251,
      "loss": 1.2621,
      "step": 155
    },
    {
      "epoch": 0.7289293849658315,
      "grad_norm": 1.5,
      "learning_rate": 0.00014611872146118722,
      "loss": 1.2575,
      "step": 160
    },
    {
      "epoch": 0.7517084282460137,
      "grad_norm": 15.1875,
      "learning_rate": 0.00015068493150684933,
      "loss": 1.2592,
      "step": 165
    },
    {
      "epoch": 0.7744874715261959,
      "grad_norm": 2.03125,
      "learning_rate": 0.0001552511415525114,
      "loss": 1.2579,
      "step": 170
    },
    {
      "epoch": 0.7972665148063781,
      "grad_norm": 1.03125,
      "learning_rate": 0.00015981735159817351,
      "loss": 1.234,
      "step": 175
    },
    {
      "epoch": 0.8200455580865603,
      "grad_norm": 3.46875,
      "learning_rate": 0.00016438356164383562,
      "loss": 1.2226,
      "step": 180
    },
    {
      "epoch": 0.8428246013667426,
      "grad_norm": 2.328125,
      "learning_rate": 0.00016894977168949773,
      "loss": 1.2259,
      "step": 185
    },
    {
      "epoch": 0.8656036446469249,
      "grad_norm": 4.25,
      "learning_rate": 0.0001735159817351598,
      "loss": 1.227,
      "step": 190
    },
    {
      "epoch": 0.8883826879271071,
      "grad_norm": 2.03125,
      "learning_rate": 0.00017808219178082192,
      "loss": 1.2009,
      "step": 195
    },
    {
      "epoch": 0.9111617312072893,
      "grad_norm": 2.53125,
      "learning_rate": 0.00018264840182648402,
      "loss": 1.1779,
      "step": 200
    },
    {
      "epoch": 0.9339407744874715,
      "grad_norm": 2.75,
      "learning_rate": 0.00018721461187214613,
      "loss": 1.185,
      "step": 205
    },
    {
      "epoch": 0.9567198177676538,
      "grad_norm": 2.015625,
      "learning_rate": 0.0001917808219178082,
      "loss": 1.168,
      "step": 210
    },
    {
      "epoch": 0.979498861047836,
      "grad_norm": 2.421875,
      "learning_rate": 0.00019634703196347032,
      "loss": 1.1594,
      "step": 215
    },
    {
      "epoch": 0.9977220956719818,
      "eval_loss": 2.6195228099823,
      "eval_runtime": 0.2786,
      "eval_samples_per_second": 35.888,
      "eval_steps_per_second": 3.589,
      "step": 219
    },
    {
      "epoch": 1.0022779043280183,
      "grad_norm": 15.5625,
      "learning_rate": 0.00019999987297289245,
      "loss": 1.165,
      "step": 220
    },
    {
      "epoch": 1.0250569476082005,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019999542705801296,
      "loss": 1.1666,
      "step": 225
    },
    {
      "epoch": 1.0478359908883828,
      "grad_norm": 4.65625,
      "learning_rate": 0.00019998463011046926,
      "loss": 1.168,
      "step": 230
    },
    {
      "epoch": 1.070615034168565,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019996748281601038,
      "loss": 1.1708,
      "step": 235
    },
    {
      "epoch": 1.0933940774487472,
      "grad_norm": 0.984375,
      "learning_rate": 0.00019994398626371643,
      "loss": 1.1528,
      "step": 240
    },
    {
      "epoch": 1.1161731207289294,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001999141419459293,
      "loss": 1.1295,
      "step": 245
    },
    {
      "epoch": 1.1389521640091116,
      "grad_norm": 1.25,
      "learning_rate": 0.00019987795175815807,
      "loss": 1.1235,
      "step": 250
    },
    {
      "epoch": 1.1617312072892938,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001998354179989585,
      "loss": 1.1176,
      "step": 255
    },
    {
      "epoch": 1.184510250569476,
      "grad_norm": 2.140625,
      "learning_rate": 0.0001997865433697871,
      "loss": 1.1074,
      "step": 260
    },
    {
      "epoch": 1.2072892938496582,
      "grad_norm": 7.53125,
      "learning_rate": 0.00019973133097482947,
      "loss": 1.1124,
      "step": 265
    },
    {
      "epoch": 1.2300683371298406,
      "grad_norm": 2.40625,
      "learning_rate": 0.00019966978432080316,
      "loss": 1.1298,
      "step": 270
    },
    {
      "epoch": 1.2528473804100229,
      "grad_norm": 2.515625,
      "learning_rate": 0.00019960190731673505,
      "loss": 1.1047,
      "step": 275
    },
    {
      "epoch": 1.275626423690205,
      "grad_norm": 1.75,
      "learning_rate": 0.00019952770427371304,
      "loss": 1.1036,
      "step": 280
    },
    {
      "epoch": 1.2984054669703873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019944717990461207,
      "loss": 1.0868,
      "step": 285
    },
    {
      "epoch": 1.3211845102505695,
      "grad_norm": 1.015625,
      "learning_rate": 0.00019936033932379504,
      "loss": 1.0943,
      "step": 290
    },
    {
      "epoch": 1.3439635535307517,
      "grad_norm": 3.046875,
      "learning_rate": 0.00019926718804678785,
      "loss": 1.098,
      "step": 295
    },
    {
      "epoch": 1.366742596810934,
      "grad_norm": 2.375,
      "learning_rate": 0.000199167731989929,
      "loss": 1.1416,
      "step": 300
    },
    {
      "epoch": 1.3895216400911161,
      "grad_norm": 4.65625,
      "learning_rate": 0.00019906197746999408,
      "loss": 1.0961,
      "step": 305
    },
    {
      "epoch": 1.4123006833712983,
      "grad_norm": 2.515625,
      "learning_rate": 0.00019894993120379435,
      "loss": 1.0799,
      "step": 310
    },
    {
      "epoch": 1.4350797266514808,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00019883160030775016,
      "loss": 1.0671,
      "step": 315
    },
    {
      "epoch": 1.4578587699316627,
      "grad_norm": 4.5625,
      "learning_rate": 0.00019870699229743911,
      "loss": 1.0786,
      "step": 320
    },
    {
      "epoch": 1.4806378132118452,
      "grad_norm": 2.03125,
      "learning_rate": 0.0001985761150871185,
      "loss": 1.0807,
      "step": 325
    },
    {
      "epoch": 1.5034168564920274,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00019843897698922284,
      "loss": 1.0898,
      "step": 330
    },
    {
      "epoch": 1.5261958997722096,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019829558671383585,
      "loss": 1.0734,
      "step": 335
    },
    {
      "epoch": 1.5489749430523918,
      "grad_norm": 2.484375,
      "learning_rate": 0.00019814595336813725,
      "loss": 1.0776,
      "step": 340
    },
    {
      "epoch": 1.571753986332574,
      "grad_norm": 2.25,
      "learning_rate": 0.0001979900864558242,
      "loss": 1.0672,
      "step": 345
    },
    {
      "epoch": 1.5945330296127562,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019782799587650805,
      "loss": 1.0553,
      "step": 350
    },
    {
      "epoch": 1.6173120728929384,
      "grad_norm": 0.671875,
      "learning_rate": 0.00019765969192508508,
      "loss": 1.0515,
      "step": 355
    },
    {
      "epoch": 1.6400911161731209,
      "grad_norm": 2.203125,
      "learning_rate": 0.00019748518529108316,
      "loss": 1.0499,
      "step": 360
    },
    {
      "epoch": 1.6628701594533029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019730448705798239,
      "loss": 1.0539,
      "step": 365
    },
    {
      "epoch": 1.6856492027334853,
      "grad_norm": 4.65625,
      "learning_rate": 0.00019711760870251143,
      "loss": 1.0605,
      "step": 370
    },
    {
      "epoch": 1.7084282460136673,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00019692456209391846,
      "loss": 1.0732,
      "step": 375
    },
    {
      "epoch": 1.7312072892938497,
      "grad_norm": 5.8125,
      "learning_rate": 0.0001967253594932173,
      "loss": 1.0485,
      "step": 380
    },
    {
      "epoch": 1.753986332574032,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019652001355240878,
      "loss": 1.0665,
      "step": 385
    },
    {
      "epoch": 1.7767653758542141,
      "grad_norm": 0.953125,
      "learning_rate": 0.00019630853731367713,
      "loss": 1.0505,
      "step": 390
    },
    {
      "epoch": 1.7995444191343963,
      "grad_norm": 8.125,
      "learning_rate": 0.0001960909442085615,
      "loss": 1.0415,
      "step": 395
    },
    {
      "epoch": 1.8223234624145785,
      "grad_norm": 2.078125,
      "learning_rate": 0.00019586724805710306,
      "loss": 1.0477,
      "step": 400
    },
    {
      "epoch": 1.845102505694761,
      "grad_norm": 2.828125,
      "learning_rate": 0.0001956374630669672,
      "loss": 1.0509,
      "step": 405
    },
    {
      "epoch": 1.867881548974943,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00019540160383254107,
      "loss": 1.0417,
      "step": 410
    },
    {
      "epoch": 1.8906605922551254,
      "grad_norm": 2.453125,
      "learning_rate": 0.00019515968533400673,
      "loss": 1.0382,
      "step": 415
    },
    {
      "epoch": 1.9134396355353074,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00019491172293638968,
      "loss": 1.0424,
      "step": 420
    },
    {
      "epoch": 1.9362186788154898,
      "grad_norm": 9.1875,
      "learning_rate": 0.00019465773238858298,
      "loss": 1.0425,
      "step": 425
    },
    {
      "epoch": 1.958997722095672,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00019439772982234697,
      "loss": 1.041,
      "step": 430
    },
    {
      "epoch": 1.9817767653758542,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019413173175128473,
      "loss": 1.0276,
      "step": 435
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.4670326709747314,
      "eval_runtime": 0.2346,
      "eval_samples_per_second": 42.634,
      "eval_steps_per_second": 4.263,
      "step": 439
    },
    {
      "epoch": 2.0045558086560367,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0001938597550697932,
      "loss": 1.0312,
      "step": 440
    },
    {
      "epoch": 2.0273348519362187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019358181705199015,
      "loss": 1.0035,
      "step": 445
    },
    {
      "epoch": 2.050113895216401,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00019329793535061723,
      "loss": 1.0027,
      "step": 450
    },
    {
      "epoch": 2.072892938496583,
      "grad_norm": 1.03125,
      "learning_rate": 0.00019300812799591846,
      "loss": 0.9951,
      "step": 455
    },
    {
      "epoch": 2.0956719817767655,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00019271241339449536,
      "loss": 1.0143,
      "step": 460
    },
    {
      "epoch": 2.1184510250569475,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00019241081032813772,
      "loss": 0.9942,
      "step": 465
    },
    {
      "epoch": 2.14123006833713,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00019210333795263075,
      "loss": 1.0064,
      "step": 470
    },
    {
      "epoch": 2.164009111617312,
      "grad_norm": 3.96875,
      "learning_rate": 0.00019179001579653853,
      "loss": 0.9909,
      "step": 475
    },
    {
      "epoch": 2.1867881548974943,
      "grad_norm": 2.359375,
      "learning_rate": 0.0001914708637599636,
      "loss": 1.006,
      "step": 480
    },
    {
      "epoch": 2.2095671981776768,
      "grad_norm": 1.375,
      "learning_rate": 0.00019114590211328288,
      "loss": 1.0048,
      "step": 485
    },
    {
      "epoch": 2.2323462414578588,
      "grad_norm": 2.828125,
      "learning_rate": 0.0001908151514958606,
      "loss": 1.0088,
      "step": 490
    },
    {
      "epoch": 2.255125284738041,
      "grad_norm": 4.0,
      "learning_rate": 0.00019047863291473717,
      "loss": 1.0095,
      "step": 495
    },
    {
      "epoch": 2.277904328018223,
      "grad_norm": 10.875,
      "learning_rate": 0.00019013636774329495,
      "loss": 1.0044,
      "step": 500
    },
    {
      "epoch": 2.3006833712984056,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018978837771990085,
      "loss": 1.0028,
      "step": 505
    },
    {
      "epoch": 2.3234624145785876,
      "grad_norm": 1.921875,
      "learning_rate": 0.0001894346849465257,
      "loss": 0.9945,
      "step": 510
    },
    {
      "epoch": 2.34624145785877,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00018907531188734026,
      "loss": 0.9848,
      "step": 515
    },
    {
      "epoch": 2.369020501138952,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018871028136728874,
      "loss": 0.9888,
      "step": 520
    },
    {
      "epoch": 2.3917995444191344,
      "grad_norm": 3.15625,
      "learning_rate": 0.00018833961657063885,
      "loss": 1.0026,
      "step": 525
    },
    {
      "epoch": 2.4145785876993164,
      "grad_norm": 11.1875,
      "learning_rate": 0.0001879633410395095,
      "loss": 1.0341,
      "step": 530
    },
    {
      "epoch": 2.437357630979499,
      "grad_norm": 2.1875,
      "learning_rate": 0.00018758147867237548,
      "loss": 1.0097,
      "step": 535
    },
    {
      "epoch": 2.4601366742596813,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018719405372254948,
      "loss": 1.0025,
      "step": 540
    },
    {
      "epoch": 2.4829157175398633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018680109079664188,
      "loss": 1.0056,
      "step": 545
    },
    {
      "epoch": 2.5056947608200457,
      "grad_norm": 3.609375,
      "learning_rate": 0.0001864026148529978,
      "loss": 1.0143,
      "step": 550
    },
    {
      "epoch": 2.5284738041002277,
      "grad_norm": 0.76171875,
      "learning_rate": 0.00018599865120011192,
      "loss": 1.0024,
      "step": 555
    },
    {
      "epoch": 2.55125284738041,
      "grad_norm": 17.875,
      "learning_rate": 0.00018558922549502107,
      "loss": 1.0595,
      "step": 560
    },
    {
      "epoch": 2.574031890660592,
      "grad_norm": 1.0,
      "learning_rate": 0.0001851743637416747,
      "loss": 1.0061,
      "step": 565
    },
    {
      "epoch": 2.5968109339407746,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018475409228928312,
      "loss": 0.9936,
      "step": 570
    },
    {
      "epoch": 2.619589977220957,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018432843783064429,
      "loss": 1.0232,
      "step": 575
    },
    {
      "epoch": 2.642369020501139,
      "grad_norm": 5.375,
      "learning_rate": 0.00018389742740044813,
      "loss": 0.989,
      "step": 580
    },
    {
      "epoch": 2.665148063781321,
      "grad_norm": 3.671875,
      "learning_rate": 0.00018346108837355972,
      "loss": 0.9927,
      "step": 585
    },
    {
      "epoch": 2.6879271070615034,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00018301944846328049,
      "loss": 0.9807,
      "step": 590
    },
    {
      "epoch": 2.710706150341686,
      "grad_norm": 0.70703125,
      "learning_rate": 0.0001825725357195881,
      "loss": 1.0005,
      "step": 595
    },
    {
      "epoch": 2.733485193621868,
      "grad_norm": 1.09375,
      "learning_rate": 0.00018212037852735486,
      "loss": 0.9641,
      "step": 600
    },
    {
      "epoch": 2.7562642369020502,
      "grad_norm": 1.875,
      "learning_rate": 0.0001816630056045451,
      "loss": 0.9897,
      "step": 605
    },
    {
      "epoch": 2.7790432801822322,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0001812004460003909,
      "loss": 0.9767,
      "step": 610
    },
    {
      "epoch": 2.8018223234624147,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00018073272909354727,
      "loss": 0.9739,
      "step": 615
    },
    {
      "epoch": 2.8246013667425967,
      "grad_norm": 2.90625,
      "learning_rate": 0.0001802598845902262,
      "loss": 0.9698,
      "step": 620
    },
    {
      "epoch": 2.847380410022779,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017978194252230985,
      "loss": 0.9707,
      "step": 625
    },
    {
      "epoch": 2.8701594533029615,
      "grad_norm": 2.46875,
      "learning_rate": 0.00017929893324544332,
      "loss": 0.9717,
      "step": 630
    },
    {
      "epoch": 2.8929384965831435,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001788108874371063,
      "loss": 0.9655,
      "step": 635
    },
    {
      "epoch": 2.9157175398633255,
      "grad_norm": 2.453125,
      "learning_rate": 0.00017831783609466504,
      "loss": 0.998,
      "step": 640
    },
    {
      "epoch": 2.938496583143508,
      "grad_norm": 1.828125,
      "learning_rate": 0.00017781981053340337,
      "loss": 0.9503,
      "step": 645
    },
    {
      "epoch": 2.9612756264236904,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00017731684238453385,
      "loss": 0.9531,
      "step": 650
    },
    {
      "epoch": 2.9840546697038723,
      "grad_norm": 0.875,
      "learning_rate": 0.0001768089635931887,
      "loss": 0.9492,
      "step": 655
    },
    {
      "epoch": 2.997722095671982,
      "eval_loss": 2.445070266723633,
      "eval_runtime": 0.2582,
      "eval_samples_per_second": 38.725,
      "eval_steps_per_second": 3.872,
      "step": 658
    },
    {
      "epoch": 3.0068337129840548,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00017629620641639103,
      "loss": 0.9628,
      "step": 660
    },
    {
      "epoch": 3.0296127562642368,
      "grad_norm": 0.90625,
      "learning_rate": 0.00017577860342100579,
      "loss": 0.9402,
      "step": 665
    },
    {
      "epoch": 3.052391799544419,
      "grad_norm": 0.6875,
      "learning_rate": 0.0001752561874816717,
      "loss": 0.9384,
      "step": 670
    },
    {
      "epoch": 3.075170842824601,
      "grad_norm": 1.8125,
      "learning_rate": 0.00017472899177871297,
      "loss": 0.9494,
      "step": 675
    },
    {
      "epoch": 3.0979498861047836,
      "grad_norm": 2.765625,
      "learning_rate": 0.00017419704979603214,
      "loss": 0.9598,
      "step": 680
    },
    {
      "epoch": 3.120728929384966,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017366039531898326,
      "loss": 0.9588,
      "step": 685
    },
    {
      "epoch": 3.143507972665148,
      "grad_norm": 11.5625,
      "learning_rate": 0.00017311906243222614,
      "loss": 0.9401,
      "step": 690
    },
    {
      "epoch": 3.1662870159453305,
      "grad_norm": 7.125,
      "learning_rate": 0.0001725730855175615,
      "loss": 0.9367,
      "step": 695
    },
    {
      "epoch": 3.1890660592255125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017202249925174723,
      "loss": 0.932,
      "step": 700
    },
    {
      "epoch": 3.211845102505695,
      "grad_norm": 3.65625,
      "learning_rate": 0.00017146733860429612,
      "loss": 0.9413,
      "step": 705
    },
    {
      "epoch": 3.234624145785877,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001709076388352546,
      "loss": 0.95,
      "step": 710
    },
    {
      "epoch": 3.2574031890660593,
      "grad_norm": 18.0,
      "learning_rate": 0.00017034343549296346,
      "loss": 0.9519,
      "step": 715
    },
    {
      "epoch": 3.2801822323462413,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016977476441179992,
      "loss": 0.932,
      "step": 720
    },
    {
      "epoch": 3.3029612756264237,
      "grad_norm": 0.71484375,
      "learning_rate": 0.0001692016617099018,
      "loss": 0.9213,
      "step": 725
    },
    {
      "epoch": 3.3257403189066057,
      "grad_norm": 2.09375,
      "learning_rate": 0.0001686241637868734,
      "loss": 0.9203,
      "step": 730
    },
    {
      "epoch": 3.348519362186788,
      "grad_norm": 3.03125,
      "learning_rate": 0.0001680423073214737,
      "loss": 0.9136,
      "step": 735
    },
    {
      "epoch": 3.3712984054669706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016745612926928694,
      "loss": 0.9299,
      "step": 740
    },
    {
      "epoch": 3.3940774487471526,
      "grad_norm": 2.203125,
      "learning_rate": 0.0001668656668603751,
      "loss": 0.9088,
      "step": 745
    },
    {
      "epoch": 3.416856492027335,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00016627095759691362,
      "loss": 0.9212,
      "step": 750
    },
    {
      "epoch": 3.439635535307517,
      "grad_norm": 2.015625,
      "learning_rate": 0.0001656720392508094,
      "loss": 0.9136,
      "step": 755
    },
    {
      "epoch": 3.4624145785876994,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00016506894986130171,
      "loss": 0.9163,
      "step": 760
    },
    {
      "epoch": 3.4851936218678814,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00016446172773254629,
      "loss": 0.917,
      "step": 765
    },
    {
      "epoch": 3.507972665148064,
      "grad_norm": 5.21875,
      "learning_rate": 0.00016385041143118255,
      "loss": 0.9071,
      "step": 770
    },
    {
      "epoch": 3.5307517084282463,
      "grad_norm": 0.71875,
      "learning_rate": 0.000163235039783884,
      "loss": 0.897,
      "step": 775
    },
    {
      "epoch": 3.5535307517084282,
      "grad_norm": 0.6953125,
      "learning_rate": 0.0001626156518748922,
      "loss": 0.8978,
      "step": 780
    },
    {
      "epoch": 3.5763097949886102,
      "grad_norm": 0.498046875,
      "learning_rate": 0.00016199228704353455,
      "loss": 0.896,
      "step": 785
    },
    {
      "epoch": 3.5990888382687927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016136498488172568,
      "loss": 0.8931,
      "step": 790
    },
    {
      "epoch": 3.621867881548975,
      "grad_norm": 0.59375,
      "learning_rate": 0.0001607337852314527,
      "loss": 0.8912,
      "step": 795
    },
    {
      "epoch": 3.644646924829157,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00016009872818224485,
      "loss": 0.892,
      "step": 800
    },
    {
      "epoch": 3.6674259681093395,
      "grad_norm": 0.66015625,
      "learning_rate": 0.00015945985406862721,
      "loss": 0.9028,
      "step": 805
    },
    {
      "epoch": 3.6902050113895215,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00015881720346755905,
      "loss": 0.8992,
      "step": 810
    },
    {
      "epoch": 3.712984054669704,
      "grad_norm": 0.61328125,
      "learning_rate": 0.00015817081719585643,
      "loss": 0.9005,
      "step": 815
    },
    {
      "epoch": 3.735763097949886,
      "grad_norm": 16.625,
      "learning_rate": 0.00015752073630759998,
      "loss": 0.9018,
      "step": 820
    },
    {
      "epoch": 3.7585421412300684,
      "grad_norm": 0.71484375,
      "learning_rate": 0.00015686700209152738,
      "loss": 0.9036,
      "step": 825
    },
    {
      "epoch": 3.781321184510251,
      "grad_norm": 0.69921875,
      "learning_rate": 0.00015620965606841098,
      "loss": 0.8872,
      "step": 830
    },
    {
      "epoch": 3.8041002277904328,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001555487399884206,
      "loss": 0.8847,
      "step": 835
    },
    {
      "epoch": 3.8268792710706148,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00015488429582847192,
      "loss": 0.8814,
      "step": 840
    },
    {
      "epoch": 3.849658314350797,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0001542163657895605,
      "loss": 0.8898,
      "step": 845
    },
    {
      "epoch": 3.8724373576309796,
      "grad_norm": 3.34375,
      "learning_rate": 0.00015354499229408114,
      "loss": 0.8841,
      "step": 850
    },
    {
      "epoch": 3.8952164009111616,
      "grad_norm": 0.75,
      "learning_rate": 0.0001528702179831338,
      "loss": 0.879,
      "step": 855
    },
    {
      "epoch": 3.917995444191344,
      "grad_norm": 0.64453125,
      "learning_rate": 0.00015219208571381525,
      "loss": 0.8911,
      "step": 860
    },
    {
      "epoch": 3.940774487471526,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00015151063855649698,
      "loss": 0.8844,
      "step": 865
    },
    {
      "epoch": 3.9635535307517085,
      "grad_norm": 0.51953125,
      "learning_rate": 0.00015082591979208976,
      "loss": 0.8787,
      "step": 870
    },
    {
      "epoch": 3.9863325740318905,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00015013797290929466,
      "loss": 0.8751,
      "step": 875
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.435899496078491,
      "eval_runtime": 0.2346,
      "eval_samples_per_second": 42.632,
      "eval_steps_per_second": 4.263,
      "step": 878
    },
    {
      "epoch": 4.009111617312073,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00014944684160184108,
      "loss": 0.8723,
      "step": 880
    },
    {
      "epoch": 4.031890660592255,
      "grad_norm": 0.44921875,
      "learning_rate": 0.00014875256976571135,
      "loss": 0.8563,
      "step": 885
    },
    {
      "epoch": 4.054669703872437,
      "grad_norm": 0.546875,
      "learning_rate": 0.00014805520149635307,
      "loss": 0.8632,
      "step": 890
    },
    {
      "epoch": 4.077448747152619,
      "grad_norm": 0.53125,
      "learning_rate": 0.00014735478108587828,
      "loss": 0.8585,
      "step": 895
    },
    {
      "epoch": 4.100227790432802,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00014665135302025035,
      "loss": 0.8662,
      "step": 900
    },
    {
      "epoch": 4.123006833712984,
      "grad_norm": 0.609375,
      "learning_rate": 0.00014594496197645852,
      "loss": 0.8569,
      "step": 905
    },
    {
      "epoch": 4.145785876993166,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0001452356528196804,
      "loss": 0.8647,
      "step": 910
    },
    {
      "epoch": 4.168564920273348,
      "grad_norm": 0.76171875,
      "learning_rate": 0.00014452347060043237,
      "loss": 0.8513,
      "step": 915
    },
    {
      "epoch": 4.191343963553531,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00014380846055170828,
      "loss": 0.8568,
      "step": 920
    },
    {
      "epoch": 4.214123006833713,
      "grad_norm": 0.640625,
      "learning_rate": 0.00014309066808610655,
      "loss": 0.8645,
      "step": 925
    },
    {
      "epoch": 4.236902050113895,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001423701387929459,
      "loss": 0.8526,
      "step": 930
    },
    {
      "epoch": 4.259681093394078,
      "grad_norm": 0.490234375,
      "learning_rate": 0.00014164691843536982,
      "loss": 0.8656,
      "step": 935
    },
    {
      "epoch": 4.28246013667426,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00014092105294744,
      "loss": 0.8579,
      "step": 940
    },
    {
      "epoch": 4.305239179954442,
      "grad_norm": 0.75390625,
      "learning_rate": 0.00014019258843121893,
      "loss": 0.854,
      "step": 945
    },
    {
      "epoch": 4.328018223234624,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0001394615711538417,
      "loss": 0.8517,
      "step": 950
    },
    {
      "epoch": 4.350797266514807,
      "grad_norm": 0.51953125,
      "learning_rate": 0.00013872804754457759,
      "loss": 0.8575,
      "step": 955
    },
    {
      "epoch": 4.373576309794989,
      "grad_norm": 0.58203125,
      "learning_rate": 0.00013799206419188103,
      "loss": 0.8528,
      "step": 960
    },
    {
      "epoch": 4.396355353075171,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00013725366784043288,
      "loss": 0.8551,
      "step": 965
    },
    {
      "epoch": 4.4191343963553535,
      "grad_norm": 0.62109375,
      "learning_rate": 0.00013651290538817113,
      "loss": 0.8553,
      "step": 970
    },
    {
      "epoch": 4.4419134396355355,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0001357698238833126,
      "loss": 0.8585,
      "step": 975
    },
    {
      "epoch": 4.4646924829157175,
      "grad_norm": 0.5625,
      "learning_rate": 0.00013502447052136455,
      "loss": 0.8511,
      "step": 980
    },
    {
      "epoch": 4.4874715261958995,
      "grad_norm": 0.427734375,
      "learning_rate": 0.00013427689264212738,
      "loss": 0.8612,
      "step": 985
    },
    {
      "epoch": 4.510250569476082,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00013352713772668765,
      "loss": 0.8502,
      "step": 990
    },
    {
      "epoch": 4.533029612756264,
      "grad_norm": 0.640625,
      "learning_rate": 0.0001327752533944025,
      "loss": 0.858,
      "step": 995
    },
    {
      "epoch": 4.555808656036446,
      "grad_norm": 0.447265625,
      "learning_rate": 0.00013202128739987532,
      "loss": 0.8462,
      "step": 1000
    },
    {
      "epoch": 4.578587699316628,
      "grad_norm": 0.45703125,
      "learning_rate": 0.00013126528762992247,
      "loss": 0.8539,
      "step": 1005
    },
    {
      "epoch": 4.601366742596811,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001305073021005321,
      "loss": 0.8562,
      "step": 1010
    },
    {
      "epoch": 4.624145785876993,
      "grad_norm": 0.640625,
      "learning_rate": 0.0001297473789538142,
      "loss": 0.8625,
      "step": 1015
    },
    {
      "epoch": 4.646924829157175,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00012898556645494325,
      "loss": 0.8547,
      "step": 1020
    },
    {
      "epoch": 4.669703872437358,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0001282219129890925,
      "loss": 0.849,
      "step": 1025
    },
    {
      "epoch": 4.69248291571754,
      "grad_norm": 0.41796875,
      "learning_rate": 0.00012745646705836097,
      "loss": 0.8469,
      "step": 1030
    },
    {
      "epoch": 4.715261958997722,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001266892772786929,
      "loss": 0.8512,
      "step": 1035
    },
    {
      "epoch": 4.738041002277904,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0001259203923767901,
      "loss": 0.8537,
      "step": 1040
    },
    {
      "epoch": 4.760820045558087,
      "grad_norm": 0.458984375,
      "learning_rate": 0.00012514986118701695,
      "loss": 0.8512,
      "step": 1045
    },
    {
      "epoch": 4.783599088838269,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00012437773264829897,
      "loss": 0.8507,
      "step": 1050
    },
    {
      "epoch": 4.806378132118451,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00012360405580101448,
      "loss": 0.8485,
      "step": 1055
    },
    {
      "epoch": 4.829157175398633,
      "grad_norm": 0.470703125,
      "learning_rate": 0.00012282887978387976,
      "loss": 0.8498,
      "step": 1060
    },
    {
      "epoch": 4.851936218678816,
      "grad_norm": 0.59765625,
      "learning_rate": 0.00012205225383082843,
      "loss": 0.8518,
      "step": 1065
    },
    {
      "epoch": 4.874715261958998,
      "grad_norm": 0.62109375,
      "learning_rate": 0.000121274227267884,
      "loss": 0.846,
      "step": 1070
    },
    {
      "epoch": 4.89749430523918,
      "grad_norm": 0.41015625,
      "learning_rate": 0.00012049484951002739,
      "loss": 0.8454,
      "step": 1075
    },
    {
      "epoch": 4.920273348519363,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00011971417005805818,
      "loss": 0.8457,
      "step": 1080
    },
    {
      "epoch": 4.943052391799545,
      "grad_norm": 0.59375,
      "learning_rate": 0.00011893223849545084,
      "loss": 0.845,
      "step": 1085
    },
    {
      "epoch": 4.965831435079727,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00011814910448520536,
      "loss": 0.8473,
      "step": 1090
    },
    {
      "epoch": 4.988610478359909,
      "grad_norm": 0.439453125,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.8477,
      "step": 1095
    },
    {
      "epoch": 4.997722095671982,
      "eval_loss": 2.439002275466919,
      "eval_runtime": 0.2582,
      "eval_samples_per_second": 38.729,
      "eval_steps_per_second": 3.873,
      "step": 1097
    },
    {
      "epoch": 5.011389521640091,
      "grad_norm": 0.60546875,
      "learning_rate": 0.00011657942815249754,
      "loss": 0.8327,
      "step": 1100
    },
    {
      "epoch": 5.034168564920273,
      "grad_norm": 0.50390625,
      "learning_rate": 0.00011579298552525084,
      "loss": 0.8204,
      "step": 1105
    },
    {
      "epoch": 5.056947608200455,
      "grad_norm": 0.427734375,
      "learning_rate": 0.00011500553983446527,
      "loss": 0.8223,
      "step": 1110
    },
    {
      "epoch": 5.079726651480637,
      "grad_norm": 0.46484375,
      "learning_rate": 0.00011421714109336097,
      "loss": 0.8185,
      "step": 1115
    },
    {
      "epoch": 5.10250569476082,
      "grad_norm": 0.458984375,
      "learning_rate": 0.00011342783937568926,
      "loss": 0.8175,
      "step": 1120
    },
    {
      "epoch": 5.125284738041002,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00011263768481255264,
      "loss": 0.8238,
      "step": 1125
    },
    {
      "epoch": 5.148063781321184,
      "grad_norm": 0.494140625,
      "learning_rate": 0.00011184672758922034,
      "loss": 0.8184,
      "step": 1130
    },
    {
      "epoch": 5.170842824601367,
      "grad_norm": 0.4140625,
      "learning_rate": 0.00011105501794194131,
      "loss": 0.8184,
      "step": 1135
    },
    {
      "epoch": 5.193621867881549,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00011026260615475333,
      "loss": 0.8265,
      "step": 1140
    },
    {
      "epoch": 5.216400911161731,
      "grad_norm": 0.427734375,
      "learning_rate": 0.00010946954255628928,
      "loss": 0.83,
      "step": 1145
    },
    {
      "epoch": 5.239179954441913,
      "grad_norm": 0.435546875,
      "learning_rate": 0.00010867587751658079,
      "loss": 0.8217,
      "step": 1150
    },
    {
      "epoch": 5.261958997722096,
      "grad_norm": 0.478515625,
      "learning_rate": 0.00010788166144385888,
      "loss": 0.8234,
      "step": 1155
    },
    {
      "epoch": 5.284738041002278,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0001070869447813525,
      "loss": 0.8175,
      "step": 1160
    },
    {
      "epoch": 5.30751708428246,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0001062917780040847,
      "loss": 0.8245,
      "step": 1165
    },
    {
      "epoch": 5.330296127562642,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0001054962116156667,
      "loss": 0.8147,
      "step": 1170
    },
    {
      "epoch": 5.353075170842825,
      "grad_norm": 0.41015625,
      "learning_rate": 0.00010470029614509041,
      "loss": 0.8196,
      "step": 1175
    },
    {
      "epoch": 5.375854214123007,
      "grad_norm": 0.427734375,
      "learning_rate": 0.00010390408214351892,
      "loss": 0.8208,
      "step": 1180
    },
    {
      "epoch": 5.398633257403189,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0001031076201810762,
      "loss": 0.8157,
      "step": 1185
    },
    {
      "epoch": 5.421412300683372,
      "grad_norm": 0.478515625,
      "learning_rate": 0.00010231096084363483,
      "loss": 0.8309,
      "step": 1190
    },
    {
      "epoch": 5.444191343963554,
      "grad_norm": 0.51953125,
      "learning_rate": 0.00010151415472960342,
      "loss": 0.8216,
      "step": 1195
    },
    {
      "epoch": 5.466970387243736,
      "grad_norm": 0.4140625,
      "learning_rate": 0.00010071725244671282,
      "loss": 0.824,
      "step": 1200
    },
    {
      "epoch": 5.489749430523918,
      "grad_norm": 0.5234375,
      "learning_rate": 9.992030460880181e-05,
      "loss": 0.8223,
      "step": 1205
    },
    {
      "epoch": 5.5125284738041005,
      "grad_norm": 0.51171875,
      "learning_rate": 9.91233618326026e-05,
      "loss": 0.8222,
      "step": 1210
    },
    {
      "epoch": 5.5353075170842825,
      "grad_norm": 0.4296875,
      "learning_rate": 9.83264747345259e-05,
      "loss": 0.8225,
      "step": 1215
    },
    {
      "epoch": 5.5580865603644645,
      "grad_norm": 0.458984375,
      "learning_rate": 9.752969392744606e-05,
      "loss": 0.8278,
      "step": 1220
    },
    {
      "epoch": 5.5808656036446465,
      "grad_norm": 0.4609375,
      "learning_rate": 9.673307001748661e-05,
      "loss": 0.8139,
      "step": 1225
    },
    {
      "epoch": 5.603644646924829,
      "grad_norm": 0.4765625,
      "learning_rate": 9.593665360080599e-05,
      "loss": 0.8191,
      "step": 1230
    },
    {
      "epoch": 5.626423690205011,
      "grad_norm": 0.41015625,
      "learning_rate": 9.514049526038418e-05,
      "loss": 0.8231,
      "step": 1235
    },
    {
      "epoch": 5.649202733485193,
      "grad_norm": 0.4140625,
      "learning_rate": 9.43446455628097e-05,
      "loss": 0.8228,
      "step": 1240
    },
    {
      "epoch": 5.671981776765376,
      "grad_norm": 0.435546875,
      "learning_rate": 9.354915505506839e-05,
      "loss": 0.8123,
      "step": 1245
    },
    {
      "epoch": 5.694760820045558,
      "grad_norm": 0.37890625,
      "learning_rate": 9.27540742613326e-05,
      "loss": 0.8157,
      "step": 1250
    },
    {
      "epoch": 5.71753986332574,
      "grad_norm": 0.41796875,
      "learning_rate": 9.195945367975256e-05,
      "loss": 0.8255,
      "step": 1255
    },
    {
      "epoch": 5.740318906605923,
      "grad_norm": 0.49609375,
      "learning_rate": 9.116534377924883e-05,
      "loss": 0.8224,
      "step": 1260
    },
    {
      "epoch": 5.763097949886105,
      "grad_norm": 0.5390625,
      "learning_rate": 9.037179499630703e-05,
      "loss": 0.8189,
      "step": 1265
    },
    {
      "epoch": 5.785876993166287,
      "grad_norm": 0.60546875,
      "learning_rate": 8.957885773177438e-05,
      "loss": 0.8129,
      "step": 1270
    },
    {
      "epoch": 5.808656036446469,
      "grad_norm": 0.396484375,
      "learning_rate": 8.878658234765858e-05,
      "loss": 0.8149,
      "step": 1275
    },
    {
      "epoch": 5.831435079726651,
      "grad_norm": 0.455078125,
      "learning_rate": 8.799501916392912e-05,
      "loss": 0.8173,
      "step": 1280
    },
    {
      "epoch": 5.854214123006834,
      "grad_norm": 0.3828125,
      "learning_rate": 8.720421845532151e-05,
      "loss": 0.8197,
      "step": 1285
    },
    {
      "epoch": 5.876993166287016,
      "grad_norm": 0.5078125,
      "learning_rate": 8.641423044814374e-05,
      "loss": 0.8122,
      "step": 1290
    },
    {
      "epoch": 5.899772209567198,
      "grad_norm": 0.486328125,
      "learning_rate": 8.562510531708677e-05,
      "loss": 0.8124,
      "step": 1295
    },
    {
      "epoch": 5.922551252847381,
      "grad_norm": 0.546875,
      "learning_rate": 8.48368931820373e-05,
      "loss": 0.8148,
      "step": 1300
    },
    {
      "epoch": 5.945330296127563,
      "grad_norm": 0.447265625,
      "learning_rate": 8.404964410489485e-05,
      "loss": 0.8178,
      "step": 1305
    },
    {
      "epoch": 5.968109339407745,
      "grad_norm": 0.443359375,
      "learning_rate": 8.32634080863919e-05,
      "loss": 0.808,
      "step": 1310
    },
    {
      "epoch": 5.990888382687928,
      "grad_norm": 0.486328125,
      "learning_rate": 8.247823506291844e-05,
      "loss": 0.809,
      "step": 1315
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.4545505046844482,
      "eval_runtime": 0.2415,
      "eval_samples_per_second": 41.413,
      "eval_steps_per_second": 4.141,
      "step": 1317
    },
    {
      "epoch": 6.0136674259681095,
      "grad_norm": 0.55078125,
      "learning_rate": 8.169417490335007e-05,
      "loss": 0.8048,
      "step": 1320
    },
    {
      "epoch": 6.0364464692482915,
      "grad_norm": 0.59765625,
      "learning_rate": 8.091127740588094e-05,
      "loss": 0.7976,
      "step": 1325
    },
    {
      "epoch": 6.0592255125284735,
      "grad_norm": 0.40625,
      "learning_rate": 8.012959229486061e-05,
      "loss": 0.7858,
      "step": 1330
    },
    {
      "epoch": 6.082004555808656,
      "grad_norm": 0.408203125,
      "learning_rate": 7.934916921763628e-05,
      "loss": 0.7997,
      "step": 1335
    },
    {
      "epoch": 6.104783599088838,
      "grad_norm": 0.41796875,
      "learning_rate": 7.857005774139907e-05,
      "loss": 0.8061,
      "step": 1340
    },
    {
      "epoch": 6.12756264236902,
      "grad_norm": 0.404296875,
      "learning_rate": 7.779230735003628e-05,
      "loss": 0.7975,
      "step": 1345
    },
    {
      "epoch": 6.150341685649202,
      "grad_norm": 0.416015625,
      "learning_rate": 7.701596744098818e-05,
      "loss": 0.7943,
      "step": 1350
    },
    {
      "epoch": 6.173120728929385,
      "grad_norm": 0.85546875,
      "learning_rate": 7.624108732211081e-05,
      "loss": 0.7965,
      "step": 1355
    },
    {
      "epoch": 6.195899772209567,
      "grad_norm": 0.4140625,
      "learning_rate": 7.54677162085442e-05,
      "loss": 0.7944,
      "step": 1360
    },
    {
      "epoch": 6.218678815489749,
      "grad_norm": 0.400390625,
      "learning_rate": 7.469590321958662e-05,
      "loss": 0.7987,
      "step": 1365
    },
    {
      "epoch": 6.241457858769932,
      "grad_norm": 0.431640625,
      "learning_rate": 7.392569737557474e-05,
      "loss": 0.8024,
      "step": 1370
    },
    {
      "epoch": 6.264236902050114,
      "grad_norm": 0.431640625,
      "learning_rate": 7.31571475947703e-05,
      "loss": 0.7931,
      "step": 1375
    },
    {
      "epoch": 6.287015945330296,
      "grad_norm": 0.435546875,
      "learning_rate": 7.239030269025311e-05,
      "loss": 0.8034,
      "step": 1380
    },
    {
      "epoch": 6.309794988610478,
      "grad_norm": 0.412109375,
      "learning_rate": 7.162521136682085e-05,
      "loss": 0.7963,
      "step": 1385
    },
    {
      "epoch": 6.332574031890661,
      "grad_norm": 0.474609375,
      "learning_rate": 7.08619222178954e-05,
      "loss": 0.8004,
      "step": 1390
    },
    {
      "epoch": 6.355353075170843,
      "grad_norm": 0.41015625,
      "learning_rate": 7.010048372243698e-05,
      "loss": 0.7979,
      "step": 1395
    },
    {
      "epoch": 6.378132118451025,
      "grad_norm": 0.4296875,
      "learning_rate": 6.934094424186459e-05,
      "loss": 0.799,
      "step": 1400
    },
    {
      "epoch": 6.400911161731207,
      "grad_norm": 0.453125,
      "learning_rate": 6.858335201698485e-05,
      "loss": 0.7999,
      "step": 1405
    },
    {
      "epoch": 6.42369020501139,
      "grad_norm": 0.53125,
      "learning_rate": 6.782775516492771e-05,
      "loss": 0.798,
      "step": 1410
    },
    {
      "epoch": 6.446469248291572,
      "grad_norm": 0.435546875,
      "learning_rate": 6.70742016760907e-05,
      "loss": 0.7977,
      "step": 1415
    },
    {
      "epoch": 6.469248291571754,
      "grad_norm": 0.4609375,
      "learning_rate": 6.632273941109064e-05,
      "loss": 0.8018,
      "step": 1420
    },
    {
      "epoch": 6.492027334851937,
      "grad_norm": 0.416015625,
      "learning_rate": 6.5573416097724e-05,
      "loss": 0.7899,
      "step": 1425
    },
    {
      "epoch": 6.514806378132119,
      "grad_norm": 0.443359375,
      "learning_rate": 6.482627932793553e-05,
      "loss": 0.7987,
      "step": 1430
    },
    {
      "epoch": 6.537585421412301,
      "grad_norm": 0.396484375,
      "learning_rate": 6.408137655479554e-05,
      "loss": 0.799,
      "step": 1435
    },
    {
      "epoch": 6.560364464692483,
      "grad_norm": 0.486328125,
      "learning_rate": 6.333875508948593e-05,
      "loss": 0.7986,
      "step": 1440
    },
    {
      "epoch": 6.5831435079726655,
      "grad_norm": 0.40234375,
      "learning_rate": 6.259846209829551e-05,
      "loss": 0.7912,
      "step": 1445
    },
    {
      "epoch": 6.605922551252847,
      "grad_norm": 0.51171875,
      "learning_rate": 6.186054459962399e-05,
      "loss": 0.7962,
      "step": 1450
    },
    {
      "epoch": 6.628701594533029,
      "grad_norm": 0.41015625,
      "learning_rate": 6.112504946099604e-05,
      "loss": 0.7917,
      "step": 1455
    },
    {
      "epoch": 6.651480637813211,
      "grad_norm": 0.40625,
      "learning_rate": 6.039202339608432e-05,
      "loss": 0.7961,
      "step": 1460
    },
    {
      "epoch": 6.674259681093394,
      "grad_norm": 0.515625,
      "learning_rate": 5.966151296174268e-05,
      "loss": 0.7991,
      "step": 1465
    },
    {
      "epoch": 6.697038724373576,
      "grad_norm": 0.40625,
      "learning_rate": 5.8933564555049105e-05,
      "loss": 0.7883,
      "step": 1470
    },
    {
      "epoch": 6.719817767653758,
      "grad_norm": 0.451171875,
      "learning_rate": 5.820822441035899e-05,
      "loss": 0.7985,
      "step": 1475
    },
    {
      "epoch": 6.742596810933941,
      "grad_norm": 0.4375,
      "learning_rate": 5.7485538596368496e-05,
      "loss": 0.798,
      "step": 1480
    },
    {
      "epoch": 6.765375854214123,
      "grad_norm": 0.376953125,
      "learning_rate": 5.6765553013188766e-05,
      "loss": 0.7969,
      "step": 1485
    },
    {
      "epoch": 6.788154897494305,
      "grad_norm": 0.5078125,
      "learning_rate": 5.6048313389430484e-05,
      "loss": 0.7892,
      "step": 1490
    },
    {
      "epoch": 6.810933940774487,
      "grad_norm": 0.41015625,
      "learning_rate": 5.533386527929962e-05,
      "loss": 0.795,
      "step": 1495
    },
    {
      "epoch": 6.83371298405467,
      "grad_norm": 0.404296875,
      "learning_rate": 5.462225405970401e-05,
      "loss": 0.796,
      "step": 1500
    },
    {
      "epoch": 6.856492027334852,
      "grad_norm": 0.416015625,
      "learning_rate": 5.391352492737157e-05,
      "loss": 0.7919,
      "step": 1505
    },
    {
      "epoch": 6.879271070615034,
      "grad_norm": 0.546875,
      "learning_rate": 5.3207722895979406e-05,
      "loss": 0.7903,
      "step": 1510
    },
    {
      "epoch": 6.902050113895216,
      "grad_norm": 0.5859375,
      "learning_rate": 5.2504892793295e-05,
      "loss": 0.795,
      "step": 1515
    },
    {
      "epoch": 6.924829157175399,
      "grad_norm": 0.4375,
      "learning_rate": 5.1805079258329056e-05,
      "loss": 0.7927,
      "step": 1520
    },
    {
      "epoch": 6.947608200455581,
      "grad_norm": 0.40625,
      "learning_rate": 5.110832673850039e-05,
      "loss": 0.7938,
      "step": 1525
    },
    {
      "epoch": 6.970387243735763,
      "grad_norm": 0.388671875,
      "learning_rate": 5.041467948681269e-05,
      "loss": 0.8044,
      "step": 1530
    },
    {
      "epoch": 6.993166287015946,
      "grad_norm": 0.38671875,
      "learning_rate": 4.9724181559044234e-05,
      "loss": 0.7918,
      "step": 1535
    },
    {
      "epoch": 6.997722095671982,
      "eval_loss": 2.4591777324676514,
      "eval_runtime": 0.2545,
      "eval_samples_per_second": 39.288,
      "eval_steps_per_second": 3.929,
      "step": 1536
    },
    {
      "epoch": 7.015945330296128,
      "grad_norm": 0.474609375,
      "learning_rate": 4.903687681094942e-05,
      "loss": 0.7889,
      "step": 1540
    },
    {
      "epoch": 7.03872437357631,
      "grad_norm": 0.423828125,
      "learning_rate": 4.835280889547351e-05,
      "loss": 0.7757,
      "step": 1545
    },
    {
      "epoch": 7.061503416856492,
      "grad_norm": 0.494140625,
      "learning_rate": 4.767202125998005e-05,
      "loss": 0.7753,
      "step": 1550
    },
    {
      "epoch": 7.0842824601366745,
      "grad_norm": 0.396484375,
      "learning_rate": 4.699455714349152e-05,
      "loss": 0.7781,
      "step": 1555
    },
    {
      "epoch": 7.1070615034168565,
      "grad_norm": 0.3984375,
      "learning_rate": 4.6320459573942856e-05,
      "loss": 0.7856,
      "step": 1560
    },
    {
      "epoch": 7.1298405466970385,
      "grad_norm": 0.46875,
      "learning_rate": 4.564977136544873e-05,
      "loss": 0.7768,
      "step": 1565
    },
    {
      "epoch": 7.152619589977221,
      "grad_norm": 0.392578125,
      "learning_rate": 4.49825351155843e-05,
      "loss": 0.7854,
      "step": 1570
    },
    {
      "epoch": 7.175398633257403,
      "grad_norm": 0.421875,
      "learning_rate": 4.431879320267972e-05,
      "loss": 0.789,
      "step": 1575
    },
    {
      "epoch": 7.198177676537585,
      "grad_norm": 0.42578125,
      "learning_rate": 4.3658587783128425e-05,
      "loss": 0.7933,
      "step": 1580
    },
    {
      "epoch": 7.220956719817767,
      "grad_norm": 0.396484375,
      "learning_rate": 4.300196078870982e-05,
      "loss": 0.7788,
      "step": 1585
    },
    {
      "epoch": 7.24373576309795,
      "grad_norm": 0.390625,
      "learning_rate": 4.2348953923925916e-05,
      "loss": 0.786,
      "step": 1590
    },
    {
      "epoch": 7.266514806378132,
      "grad_norm": 0.388671875,
      "learning_rate": 4.16996086633526e-05,
      "loss": 0.7851,
      "step": 1595
    },
    {
      "epoch": 7.289293849658314,
      "grad_norm": 0.4375,
      "learning_rate": 4.105396624900538e-05,
      "loss": 0.7775,
      "step": 1600
    },
    {
      "epoch": 7.312072892938497,
      "grad_norm": 0.39453125,
      "learning_rate": 4.041206768772022e-05,
      "loss": 0.7796,
      "step": 1605
    },
    {
      "epoch": 7.334851936218679,
      "grad_norm": 0.39453125,
      "learning_rate": 3.977395374854871e-05,
      "loss": 0.7814,
      "step": 1610
    },
    {
      "epoch": 7.357630979498861,
      "grad_norm": 0.3828125,
      "learning_rate": 3.913966496016891e-05,
      "loss": 0.7798,
      "step": 1615
    },
    {
      "epoch": 7.380410022779043,
      "grad_norm": 0.408203125,
      "learning_rate": 3.850924160831115e-05,
      "loss": 0.7928,
      "step": 1620
    },
    {
      "epoch": 7.403189066059226,
      "grad_norm": 0.390625,
      "learning_rate": 3.788272373319955e-05,
      "loss": 0.7922,
      "step": 1625
    },
    {
      "epoch": 7.425968109339408,
      "grad_norm": 0.404296875,
      "learning_rate": 3.726015112700859e-05,
      "loss": 0.7822,
      "step": 1630
    },
    {
      "epoch": 7.44874715261959,
      "grad_norm": 0.38671875,
      "learning_rate": 3.6641563331336125e-05,
      "loss": 0.7845,
      "step": 1635
    },
    {
      "epoch": 7.471526195899772,
      "grad_norm": 0.396484375,
      "learning_rate": 3.6026999634691725e-05,
      "loss": 0.7817,
      "step": 1640
    },
    {
      "epoch": 7.494305239179955,
      "grad_norm": 0.408203125,
      "learning_rate": 3.541649907000147e-05,
      "loss": 0.7746,
      "step": 1645
    },
    {
      "epoch": 7.517084282460137,
      "grad_norm": 0.38671875,
      "learning_rate": 3.4810100412128747e-05,
      "loss": 0.7886,
      "step": 1650
    },
    {
      "epoch": 7.539863325740319,
      "grad_norm": 0.3984375,
      "learning_rate": 3.42078421754117e-05,
      "loss": 0.7861,
      "step": 1655
    },
    {
      "epoch": 7.562642369020502,
      "grad_norm": 0.408203125,
      "learning_rate": 3.360976261121684e-05,
      "loss": 0.7778,
      "step": 1660
    },
    {
      "epoch": 7.585421412300684,
      "grad_norm": 0.384765625,
      "learning_rate": 3.3015899705509734e-05,
      "loss": 0.7829,
      "step": 1665
    },
    {
      "epoch": 7.6082004555808656,
      "grad_norm": 0.37890625,
      "learning_rate": 3.242629117644229e-05,
      "loss": 0.7799,
      "step": 1670
    },
    {
      "epoch": 7.6309794988610475,
      "grad_norm": 0.388671875,
      "learning_rate": 3.184097447195732e-05,
      "loss": 0.7734,
      "step": 1675
    },
    {
      "epoch": 7.65375854214123,
      "grad_norm": 0.40625,
      "learning_rate": 3.125998676740987e-05,
      "loss": 0.7796,
      "step": 1680
    },
    {
      "epoch": 7.676537585421412,
      "grad_norm": 0.3984375,
      "learning_rate": 3.068336496320631e-05,
      "loss": 0.787,
      "step": 1685
    },
    {
      "epoch": 7.699316628701594,
      "grad_norm": 0.390625,
      "learning_rate": 3.0111145682460507e-05,
      "loss": 0.7842,
      "step": 1690
    },
    {
      "epoch": 7.722095671981776,
      "grad_norm": 0.408203125,
      "learning_rate": 2.9543365268667867e-05,
      "loss": 0.7885,
      "step": 1695
    },
    {
      "epoch": 7.744874715261959,
      "grad_norm": 0.392578125,
      "learning_rate": 2.8980059783396953e-05,
      "loss": 0.7773,
      "step": 1700
    },
    {
      "epoch": 7.767653758542141,
      "grad_norm": 0.427734375,
      "learning_rate": 2.8421265003999286e-05,
      "loss": 0.785,
      "step": 1705
    },
    {
      "epoch": 7.790432801822323,
      "grad_norm": 0.4296875,
      "learning_rate": 2.7867016421336776e-05,
      "loss": 0.7875,
      "step": 1710
    },
    {
      "epoch": 7.813211845102506,
      "grad_norm": 0.3828125,
      "learning_rate": 2.7317349237527724e-05,
      "loss": 0.7884,
      "step": 1715
    },
    {
      "epoch": 7.835990888382688,
      "grad_norm": 0.40625,
      "learning_rate": 2.6772298363710956e-05,
      "loss": 0.7823,
      "step": 1720
    },
    {
      "epoch": 7.85876993166287,
      "grad_norm": 0.404296875,
      "learning_rate": 2.6231898417828603e-05,
      "loss": 0.7754,
      "step": 1725
    },
    {
      "epoch": 7.881548974943052,
      "grad_norm": 0.392578125,
      "learning_rate": 2.569618372242727e-05,
      "loss": 0.7713,
      "step": 1730
    },
    {
      "epoch": 7.904328018223235,
      "grad_norm": 0.431640625,
      "learning_rate": 2.5165188302478215e-05,
      "loss": 0.7828,
      "step": 1735
    },
    {
      "epoch": 7.927107061503417,
      "grad_norm": 0.39453125,
      "learning_rate": 2.4638945883216235e-05,
      "loss": 0.7912,
      "step": 1740
    },
    {
      "epoch": 7.949886104783599,
      "grad_norm": 0.388671875,
      "learning_rate": 2.411748988799769e-05,
      "loss": 0.7793,
      "step": 1745
    },
    {
      "epoch": 7.972665148063781,
      "grad_norm": 0.40234375,
      "learning_rate": 2.3600853436177672e-05,
      "loss": 0.7779,
      "step": 1750
    },
    {
      "epoch": 7.995444191343964,
      "grad_norm": 0.40625,
      "learning_rate": 2.3089069341006565e-05,
      "loss": 0.7847,
      "step": 1755
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.4783217906951904,
      "eval_runtime": 0.2354,
      "eval_samples_per_second": 42.482,
      "eval_steps_per_second": 4.248,
      "step": 1756
    },
    {
      "epoch": 8.018223234624147,
      "grad_norm": 0.37890625,
      "learning_rate": 2.2582170107545852e-05,
      "loss": 0.7742,
      "step": 1760
    },
    {
      "epoch": 8.041002277904328,
      "grad_norm": 0.4140625,
      "learning_rate": 2.2080187930603668e-05,
      "loss": 0.7806,
      "step": 1765
    },
    {
      "epoch": 8.06378132118451,
      "grad_norm": 0.412109375,
      "learning_rate": 2.1583154692689976e-05,
      "loss": 0.7805,
      "step": 1770
    },
    {
      "epoch": 8.086560364464692,
      "grad_norm": 0.427734375,
      "learning_rate": 2.109110196199171e-05,
      "loss": 0.773,
      "step": 1775
    },
    {
      "epoch": 8.109339407744875,
      "grad_norm": 0.466796875,
      "learning_rate": 2.0604060990367624e-05,
      "loss": 0.7698,
      "step": 1780
    },
    {
      "epoch": 8.132118451025057,
      "grad_norm": 0.435546875,
      "learning_rate": 2.0122062711363532e-05,
      "loss": 0.7773,
      "step": 1785
    },
    {
      "epoch": 8.154897494305239,
      "grad_norm": 0.396484375,
      "learning_rate": 1.9645137738247422e-05,
      "loss": 0.7744,
      "step": 1790
    },
    {
      "epoch": 8.177676537585421,
      "grad_norm": 0.408203125,
      "learning_rate": 1.9173316362065384e-05,
      "loss": 0.7799,
      "step": 1795
    },
    {
      "epoch": 8.200455580865604,
      "grad_norm": 0.431640625,
      "learning_rate": 1.8706628549717452e-05,
      "loss": 0.7766,
      "step": 1800
    },
    {
      "epoch": 8.223234624145785,
      "grad_norm": 0.400390625,
      "learning_rate": 1.824510394205453e-05,
      "loss": 0.7669,
      "step": 1805
    },
    {
      "epoch": 8.246013667425968,
      "grad_norm": 0.404296875,
      "learning_rate": 1.7788771851995655e-05,
      "loss": 0.7799,
      "step": 1810
    },
    {
      "epoch": 8.268792710706151,
      "grad_norm": 0.3828125,
      "learning_rate": 1.7337661262666294e-05,
      "loss": 0.7787,
      "step": 1815
    },
    {
      "epoch": 8.291571753986332,
      "grad_norm": 0.380859375,
      "learning_rate": 1.6891800825557535e-05,
      "loss": 0.7788,
      "step": 1820
    },
    {
      "epoch": 8.314350797266515,
      "grad_norm": 0.42578125,
      "learning_rate": 1.6451218858706374e-05,
      "loss": 0.7828,
      "step": 1825
    },
    {
      "epoch": 8.337129840546696,
      "grad_norm": 0.388671875,
      "learning_rate": 1.601594334489702e-05,
      "loss": 0.7714,
      "step": 1830
    },
    {
      "epoch": 8.35990888382688,
      "grad_norm": 0.380859375,
      "learning_rate": 1.5586001929883865e-05,
      "loss": 0.7842,
      "step": 1835
    },
    {
      "epoch": 8.382687927107062,
      "grad_norm": 0.38671875,
      "learning_rate": 1.516142192063521e-05,
      "loss": 0.7746,
      "step": 1840
    },
    {
      "epoch": 8.405466970387243,
      "grad_norm": 0.390625,
      "learning_rate": 1.474223028359939e-05,
      "loss": 0.7764,
      "step": 1845
    },
    {
      "epoch": 8.428246013667426,
      "grad_norm": 0.443359375,
      "learning_rate": 1.4328453642991646e-05,
      "loss": 0.7726,
      "step": 1850
    },
    {
      "epoch": 8.451025056947609,
      "grad_norm": 0.412109375,
      "learning_rate": 1.392011827910341e-05,
      "loss": 0.7766,
      "step": 1855
    },
    {
      "epoch": 8.47380410022779,
      "grad_norm": 0.38671875,
      "learning_rate": 1.3517250126632986e-05,
      "loss": 0.7757,
      "step": 1860
    },
    {
      "epoch": 8.496583143507973,
      "grad_norm": 0.42578125,
      "learning_rate": 1.311987477303842e-05,
      "loss": 0.773,
      "step": 1865
    },
    {
      "epoch": 8.519362186788156,
      "grad_norm": 0.412109375,
      "learning_rate": 1.2728017456912344e-05,
      "loss": 0.7793,
      "step": 1870
    },
    {
      "epoch": 8.542141230068337,
      "grad_norm": 0.388671875,
      "learning_rate": 1.2341703066379074e-05,
      "loss": 0.7665,
      "step": 1875
    },
    {
      "epoch": 8.56492027334852,
      "grad_norm": 0.38671875,
      "learning_rate": 1.1960956137513701e-05,
      "loss": 0.7737,
      "step": 1880
    },
    {
      "epoch": 8.5876993166287,
      "grad_norm": 0.3828125,
      "learning_rate": 1.158580085278398e-05,
      "loss": 0.7753,
      "step": 1885
    },
    {
      "epoch": 8.610478359908884,
      "grad_norm": 0.40234375,
      "learning_rate": 1.1216261039514087e-05,
      "loss": 0.7727,
      "step": 1890
    },
    {
      "epoch": 8.633257403189067,
      "grad_norm": 0.419921875,
      "learning_rate": 1.0852360168371656e-05,
      "loss": 0.7775,
      "step": 1895
    },
    {
      "epoch": 8.656036446469248,
      "grad_norm": 0.400390625,
      "learning_rate": 1.049412135187675e-05,
      "loss": 0.7831,
      "step": 1900
    },
    {
      "epoch": 8.67881548974943,
      "grad_norm": 0.380859375,
      "learning_rate": 1.0141567342934132e-05,
      "loss": 0.7757,
      "step": 1905
    },
    {
      "epoch": 8.701594533029613,
      "grad_norm": 0.400390625,
      "learning_rate": 9.794720533388024e-06,
      "loss": 0.7719,
      "step": 1910
    },
    {
      "epoch": 8.724373576309794,
      "grad_norm": 0.39453125,
      "learning_rate": 9.453602952599982e-06,
      "loss": 0.7691,
      "step": 1915
    },
    {
      "epoch": 8.747152619589977,
      "grad_norm": 0.388671875,
      "learning_rate": 9.118236266049707e-06,
      "loss": 0.7745,
      "step": 1920
    },
    {
      "epoch": 8.76993166287016,
      "grad_norm": 0.3828125,
      "learning_rate": 8.788641773959105e-06,
      "loss": 0.7774,
      "step": 1925
    },
    {
      "epoch": 8.792710706150341,
      "grad_norm": 0.400390625,
      "learning_rate": 8.464840409939267e-06,
      "loss": 0.7807,
      "step": 1930
    },
    {
      "epoch": 8.815489749430524,
      "grad_norm": 0.38671875,
      "learning_rate": 8.146852739661105e-06,
      "loss": 0.7765,
      "step": 1935
    },
    {
      "epoch": 8.838268792710707,
      "grad_norm": 0.38671875,
      "learning_rate": 7.834698959548914e-06,
      "loss": 0.7777,
      "step": 1940
    },
    {
      "epoch": 8.861047835990888,
      "grad_norm": 0.392578125,
      "learning_rate": 7.528398895497924e-06,
      "loss": 0.7732,
      "step": 1945
    },
    {
      "epoch": 8.883826879271071,
      "grad_norm": 0.3984375,
      "learning_rate": 7.2279720016148244e-06,
      "loss": 0.7691,
      "step": 1950
    },
    {
      "epoch": 8.906605922551252,
      "grad_norm": 0.388671875,
      "learning_rate": 6.933437358982409e-06,
      "loss": 0.7769,
      "step": 1955
    },
    {
      "epoch": 8.929384965831435,
      "grad_norm": 0.380859375,
      "learning_rate": 6.6448136744474474e-06,
      "loss": 0.779,
      "step": 1960
    },
    {
      "epoch": 8.952164009111618,
      "grad_norm": 0.38671875,
      "learning_rate": 6.36211927943271e-06,
      "loss": 0.7825,
      "step": 1965
    },
    {
      "epoch": 8.974943052391799,
      "grad_norm": 0.390625,
      "learning_rate": 6.085372128772637e-06,
      "loss": 0.7628,
      "step": 1970
    },
    {
      "epoch": 8.997722095671982,
      "grad_norm": 0.416015625,
      "learning_rate": 5.814589799572956e-06,
      "loss": 0.7808,
      "step": 1975
    },
    {
      "epoch": 8.997722095671982,
      "eval_loss": 2.488940477371216,
      "eval_runtime": 0.2524,
      "eval_samples_per_second": 39.612,
      "eval_steps_per_second": 3.961,
      "step": 1975
    },
    {
      "epoch": 9.020501138952165,
      "grad_norm": 0.392578125,
      "learning_rate": 5.549789490094304e-06,
      "loss": 0.7739,
      "step": 1980
    },
    {
      "epoch": 9.043280182232346,
      "grad_norm": 0.396484375,
      "learning_rate": 5.290988018659937e-06,
      "loss": 0.7769,
      "step": 1985
    },
    {
      "epoch": 9.066059225512529,
      "grad_norm": 0.396484375,
      "learning_rate": 5.038201822587474e-06,
      "loss": 0.779,
      "step": 1990
    },
    {
      "epoch": 9.08883826879271,
      "grad_norm": 0.3828125,
      "learning_rate": 4.79144695714504e-06,
      "loss": 0.7766,
      "step": 1995
    },
    {
      "epoch": 9.111617312072893,
      "grad_norm": 0.396484375,
      "learning_rate": 4.550739094531386e-06,
      "loss": 0.7772,
      "step": 2000
    },
    {
      "epoch": 9.134396355353076,
      "grad_norm": 0.38671875,
      "learning_rate": 4.316093522880648e-06,
      "loss": 0.7784,
      "step": 2005
    },
    {
      "epoch": 9.157175398633257,
      "grad_norm": 0.38671875,
      "learning_rate": 4.087525145291204e-06,
      "loss": 0.774,
      "step": 2010
    },
    {
      "epoch": 9.17995444191344,
      "grad_norm": 0.396484375,
      "learning_rate": 3.865048478879241e-06,
      "loss": 0.7727,
      "step": 2015
    },
    {
      "epoch": 9.202733485193622,
      "grad_norm": 0.3828125,
      "learning_rate": 3.6486776538566803e-06,
      "loss": 0.7746,
      "step": 2020
    },
    {
      "epoch": 9.225512528473804,
      "grad_norm": 0.40625,
      "learning_rate": 3.4384264126337328e-06,
      "loss": 0.779,
      "step": 2025
    },
    {
      "epoch": 9.248291571753986,
      "grad_norm": 0.408203125,
      "learning_rate": 3.2343081089460603e-06,
      "loss": 0.7734,
      "step": 2030
    },
    {
      "epoch": 9.27107061503417,
      "grad_norm": 0.3828125,
      "learning_rate": 3.0363357070066544e-06,
      "loss": 0.7691,
      "step": 2035
    },
    {
      "epoch": 9.29384965831435,
      "grad_norm": 0.419921875,
      "learning_rate": 2.8445217806824077e-06,
      "loss": 0.7853,
      "step": 2040
    },
    {
      "epoch": 9.316628701594533,
      "grad_norm": 0.388671875,
      "learning_rate": 2.658878512695562e-06,
      "loss": 0.7779,
      "step": 2045
    },
    {
      "epoch": 9.339407744874716,
      "grad_norm": 0.384765625,
      "learning_rate": 2.4794176938498837e-06,
      "loss": 0.7784,
      "step": 2050
    },
    {
      "epoch": 9.362186788154897,
      "grad_norm": 0.39453125,
      "learning_rate": 2.30615072228183e-06,
      "loss": 0.7721,
      "step": 2055
    },
    {
      "epoch": 9.38496583143508,
      "grad_norm": 0.392578125,
      "learning_rate": 2.139088602736616e-06,
      "loss": 0.7761,
      "step": 2060
    },
    {
      "epoch": 9.407744874715261,
      "grad_norm": 0.396484375,
      "learning_rate": 1.9782419458692193e-06,
      "loss": 0.7715,
      "step": 2065
    },
    {
      "epoch": 9.430523917995444,
      "grad_norm": 0.396484375,
      "learning_rate": 1.8236209675705274e-06,
      "loss": 0.7753,
      "step": 2070
    },
    {
      "epoch": 9.453302961275627,
      "grad_norm": 0.412109375,
      "learning_rate": 1.6752354883184717e-06,
      "loss": 0.7718,
      "step": 2075
    },
    {
      "epoch": 9.476082004555808,
      "grad_norm": 0.3984375,
      "learning_rate": 1.5330949325542797e-06,
      "loss": 0.7713,
      "step": 2080
    },
    {
      "epoch": 9.498861047835991,
      "grad_norm": 0.404296875,
      "learning_rate": 1.397208328083921e-06,
      "loss": 0.7747,
      "step": 2085
    },
    {
      "epoch": 9.521640091116174,
      "grad_norm": 0.3984375,
      "learning_rate": 1.2675843055046765e-06,
      "loss": 0.7831,
      "step": 2090
    },
    {
      "epoch": 9.544419134396355,
      "grad_norm": 0.392578125,
      "learning_rate": 1.144231097657078e-06,
      "loss": 0.7693,
      "step": 2095
    },
    {
      "epoch": 9.567198177676538,
      "grad_norm": 0.384765625,
      "learning_rate": 1.0271565391018922e-06,
      "loss": 0.766,
      "step": 2100
    },
    {
      "epoch": 9.589977220956719,
      "grad_norm": 0.38671875,
      "learning_rate": 9.163680656226303e-07,
      "loss": 0.7648,
      "step": 2105
    },
    {
      "epoch": 9.612756264236902,
      "grad_norm": 0.388671875,
      "learning_rate": 8.118727137532034e-07,
      "loss": 0.7703,
      "step": 2110
    },
    {
      "epoch": 9.635535307517085,
      "grad_norm": 0.390625,
      "learning_rate": 7.136771203310245e-07,
      "loss": 0.7743,
      "step": 2115
    },
    {
      "epoch": 9.658314350797266,
      "grad_norm": 0.3984375,
      "learning_rate": 6.21787522075512e-07,
      "loss": 0.7761,
      "step": 2120
    },
    {
      "epoch": 9.681093394077449,
      "grad_norm": 0.388671875,
      "learning_rate": 5.362097551919631e-07,
      "loss": 0.7649,
      "step": 2125
    },
    {
      "epoch": 9.703872437357631,
      "grad_norm": 0.38671875,
      "learning_rate": 4.569492550008603e-07,
      "loss": 0.7786,
      "step": 2130
    },
    {
      "epoch": 9.726651480637813,
      "grad_norm": 0.400390625,
      "learning_rate": 3.84011055592659e-07,
      "loss": 0.7718,
      "step": 2135
    },
    {
      "epoch": 9.749430523917995,
      "grad_norm": 0.380859375,
      "learning_rate": 3.1739978950806603e-07,
      "loss": 0.7755,
      "step": 2140
    },
    {
      "epoch": 9.772209567198178,
      "grad_norm": 0.40234375,
      "learning_rate": 2.5711968744382974e-07,
      "loss": 0.767,
      "step": 2145
    },
    {
      "epoch": 9.79498861047836,
      "grad_norm": 0.416015625,
      "learning_rate": 2.0317457798398888e-07,
      "loss": 0.7721,
      "step": 2150
    },
    {
      "epoch": 9.817767653758542,
      "grad_norm": 0.380859375,
      "learning_rate": 1.5556788735676675e-07,
      "loss": 0.7766,
      "step": 2155
    },
    {
      "epoch": 9.840546697038725,
      "grad_norm": 0.392578125,
      "learning_rate": 1.143026392168789e-07,
      "loss": 0.7745,
      "step": 2160
    },
    {
      "epoch": 9.863325740318906,
      "grad_norm": 0.390625,
      "learning_rate": 7.938145445357536e-08,
      "loss": 0.7757,
      "step": 2165
    },
    {
      "epoch": 9.88610478359909,
      "grad_norm": 0.408203125,
      "learning_rate": 5.0806551024129565e-08,
      "loss": 0.7783,
      "step": 2170
    },
    {
      "epoch": 9.90888382687927,
      "grad_norm": 0.388671875,
      "learning_rate": 2.8579743813006432e-08,
      "loss": 0.7669,
      "step": 2175
    },
    {
      "epoch": 9.931662870159453,
      "grad_norm": 0.40234375,
      "learning_rate": 1.270244451652136e-08,
      "loss": 0.7725,
      "step": 2180
    },
    {
      "epoch": 9.954441913439636,
      "grad_norm": 0.384765625,
      "learning_rate": 3.175661553256326e-09,
      "loss": 0.7773,
      "step": 2185
    },
    {
      "epoch": 9.977220956719817,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0,
      "loss": 0.7794,
      "step": 2190
    },
    {
      "epoch": 9.977220956719817,
      "eval_loss": 2.486895799636841,
      "eval_runtime": 0.2341,
      "eval_samples_per_second": 42.715,
      "eval_steps_per_second": 4.272,
      "step": 2190
    },
    {
      "epoch": 9.977220956719817,
      "step": 2190,
      "total_flos": 6.685995643959247e+18,
      "train_loss": 1.6744230088577967,
      "train_runtime": 5295.0828,
      "train_samples_per_second": 26.519,
      "train_steps_per_second": 0.414
    }
  ],
  "logging_steps": 5,
  "max_steps": 2190,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 6.685995643959247e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}